Details
Diff Detail
- Repository
- rG FreeBSD src repository
- Lint
Lint Skipped - Unit
Tests Skipped
Event Timeline
| lib/libsys/sysarch.x86.2 | ||
|---|---|---|
| 2–3 | You can use SPDX short form now | |
| 29 | Architecture dependent manuals have an optional architecture specifier in Dt. | |
| 46 | I think this is a subsection of DESCRIPTION. | |
| 143 | I think this is a subsection of DESCRIPTION | |
| 591 | ||
| 670–674 | AUTHORS is a standard section for this | |
| lib/libsys/sysarch.x86.2 | ||
|---|---|---|
| 29 | It must mention i386, even if i386 kernels are annihilated. There are differences in the syscall for 32bit and 64bit processes. | |
That's all from manpages mechanical side of things. But I do not really know about syscalls to review the content.
I'm not sure it makes sense to duplicate the descriptions for i386_vm86, etc. vs just referring to the other manpages. I didn't quite make it all the way through the page in terms of my grammar suggestions (which you are free to ignore). In general this looks good to me, though I wish we had a better way of dealing with having an MD manpage for sysarch(2), e.g. presumably we will one day need sysarch.aarch64 or the like.
| lib/libsys/sysarch.2 | ||
|---|---|---|
| 76 | ||
| lib/libsys/sysarch.x86.2 | ||
| 27 | ||
| 29 | How can i386 imply amd64? That makes no sense. Some i386 things (mostly kernel-y things vs userspace things) aren't implemented on amd64 and some amd64 things definitely aren't implemented on i386. And this manpage is clearly describing things that are amd64-specific so presumably it needs to list both? | |
| 30–31 | Perhaps? | |
| 37 | ||
| 39 | ||
| 52 | ||
| 54 | ||
| 55 | ||
| 63 | ||
| 72–74 | ||
| 84 | ||
| 88 | ||
| 90–91 | ||
| 92 | ||
| 103 | ||
| 109 | ||
| 112 | ||
| 114 | ||
| 129 | ||
| 132 | ||
| 142 | ||
| 147 | ||
| 149 | ||
| 150 | ||
| 151 | ||
| 153 | ||
| 162–172 | ||
| 177 | ||
| 178 | ||
| 183 | ||
| 188 | ||
| 189 | ||
| 192 | ||
| 195–196 | ||
| 198 | ||
| 205–206 | ||
| 208 | ||
| 214 | ||
| 221–222 | ||
| 224–226 | ||
| 228 | ||
| 235–236 | ||
| 242 | ||
| 247 | ||
| 248–249 | ||
| 251 | ||
| 254 | ||
| 265 | ||
| 266 | ||
| 281 | ||
| 289 | ||
| 295 | FWIW, I had a 486dx4 with VME (and a TSC and CPUID and several other Pentium things). Maybe trim the second half of this sentence? | |
| 302 | I would just s/vm86/VM86/ here throughout to be consistent. | |
| 322 | ||
| 352 | ||
| 353 | ||
| 357 | ||
| 360 | ||
| 369 | Alternatively, using a type like uint32_t might be a more concise way, e.g.: The .Fa args argument must point to a .Vt uint32_t variable that contains the new fsbase value. | |
Thank you very much for reading it so far.
Obviously I cannot get (the) articles right, I gave up on this many tens of years ago.
WRT to the i386_vm86.2 man page, I must admit that i missed it. But after reading it, I see that it less precise, in particular it documents not implemented fields. Also it misses crucial information needed to correctly enter VM86. I hesitate to remove text written by Bruce, might be I will leave it as is for now.
| lib/libsys/sysarch.x86.2 | ||
|---|---|---|
| 247 | It is not firmware, really. I used 'CPU instructions'. | |
| 295 | Well, 486dx4 appeared after Pentium were released AFAIR. But I agree that this detail is not needed. | |
Ping? If you do not intent to read it more, I do not see why the doc would sit in review.
| lib/libsys/x86/sysarch.x86.2 | ||
|---|---|---|
| 1 ↗ | (On Diff #187914) | Nit: The predominant style is a blank comment line here for consistency with C style comments. |